I will extract PDF invoice data into excel with ai
About this Gig
Tired of typing supplier invoices into Excel by hand? Send me your PDFs and get back a clean Excel file, with every number checked.
What I extract: supplier, document number, dates, due date, currency, line items, subtotal, VAT by rate, total and IBAN. Digital PDFs and scans both work.
What I check, with plain rules (no guessing): line totals vs subtotal, VAT vs rate, subtotal + VAT vs total, duplicates, IBAN checksum and missing fields. Anything that does not add up goes to a Review sheet with a clear reason.
Why me: I built an open-source AI agent that does exactly this. It has 246 automated tests and got 10/10 statuses right on 10 fictional test documents in 4 languages, in about 5.5 seconds each. It also ignores hidden instructions planted in a booby-trapped PDF.
Your files are used only for your order.
Not sure which package fits? Send me a message with a sample first.
Technology:
Excel
•
Python
Expertise:
Data extraction
•
Data manipulation
•
Data validation
My Portfolio
FAQ
Do scanned invoices work?
Yes. Digital PDFs are read directly and scanned ones are read by the AI. If a scan is too blurry to read with confidence, the invoice goes to the Review sheet instead of being guessed.
Which languages and currencies do you support?
My agent is tested on invoices in English, French, German and Italian, in CHF and EUR. For other languages, send me a sample before ordering and I will tell you honestly if it works.
What do you need from me?
Your PDF invoices, and the Excel layout you already use if you have one. If you have no layout, I deliver a clean standard one.

